U.S. patent application number 12/396145 was filed with the patent office on 2009-09-03 for adaptive bass management.
Invention is credited to Markus Christoph.
Application Number | 20090220098 12/396145 |
Document ID | / |
Family ID | 39048949 |
Filed Date | 2009-09-03 |
United States Patent
Application |
20090220098 |
Kind Code |
A1 |
Christoph; Markus |
September 3, 2009 |
ADAPTIVE BASS MANAGEMENT
Abstract
The invention relates to a method for adapting sound pressure
levels in at least one listening location, the sound pressure being
generated by a first and a second loudspeaker, each loudspeaker
having a supply channel arranged upstream thereto, where at least
the supply channel of the second loudspeaker modifies the phase of
an audio signal transmitted therethrough according to a phase
function. The method includes supplying an audio signal to the
supply channels and thus generating an acoustic sound signal;
measuring the acoustic sound signal at each listening location and
providing corresponding electrical signals representing the
measured acoustic sound signal; estimating updated transfer
characteristics for each pair of loudspeaker and listening
location; calculating an optimum offset phase function based on a
mathematical model using the estimated transfer characteristics;
updating the phase function by superposing the optimal offset phase
function thereto.
Inventors: |
Christoph; Markus;
(Straubing, DE) |
Correspondence
Address: |
O''Shea Getz P.C.
1500 MAIN ST. SUITE 912
SPRINGFIELD
MA
01115
US
|
Family ID: |
39048949 |
Appl. No.: |
12/396145 |
Filed: |
March 2, 2009 |
Current U.S.
Class: |
381/59 ;
381/103 |
Current CPC
Class: |
H04R 3/04 20130101; H04R
2499/13 20130101; H04S 7/302 20130101 |
Class at
Publication: |
381/59 ;
381/103 |
International
Class: |
H04R 29/00 20060101
H04R029/00 |
Foreign Application Data
Date |
Code |
Application Number |
Feb 28, 2008 |
EP |
08 003 731.0 |
Claims
1. A method for adapting sound pressure levels in at least one
listening location, the sound pressure being generated by first and
second loudspeakers, each loudspeaker having a supply channel
arranged upstream thereto, where at least the supply channel of the
second loudspeaker modifies the phase of an audio signal
transmitted therethrough according to a phase function, the method
comprising: supplying an audio signal to the supply channels and
generating an acoustic sound signal; measuring the acoustic sound
signal at the listening locations and for each listening location
providing corresponding electrical signals representing the
measured acoustic sound signal; estimating updated transfer
characteristics for each pair of loudspeaker and listening
location; calculating an optimum offset phase value based on a
mathematical model using the estimated transfer characteristics;
and updating the phase function by superposing the optimal offset
phase function thereto.
2. The method of claim 1, where the calculating step comprises:
simulating, for different frequencies and phase shifts in the
supply channel of the second loudspeaker, sound pressure levels at
each of the listening locations, where the phase shifts of the
audio signals supplied to the other loudspeakers are zero or
constant; evaluating, for the different frequencies and phase
shifts, a cost function dependent on the sound pressure level; and
searching a frequency dependent optimal phase shift that yields an
extremum of the cost function, thus obtaining a phase function
representing the optimal phase shift as a function of
frequency.
3. The method of claim 2, where the searching step comprises:
evaluating the cost function for pairs of phase shift and
frequency; and searching, for each frequency for which the cost
function has been evaluated, an optimal phase shift that yields an
extremum of the cost function.
4. The method of claim 2, where the cost function is dependent on
the sound pressure level, and, in the searching step, an optimal
phase shift is determined that maximizes the cost function yielding
a maximal sound pressure level.
5. The method of claim 2, where the cost function is dependent on
the sound pressure level and a reference sound pressure level, and
in the searching step, an optimal phase shift is determined using
the cost function, the cost function representing the distance
between the sound pressure level at the at least one listening
location and the reference sound pressure level.
6. The method of claim 5, where the reference sound pressure level
is a predefined target function of a desired sound pressure level
over frequency.
7. The method of claim 5, where the sound pressure levels are
calculated for at least two listening locations, and the reference
sound pressure level is either the sound pressure level calculated
for the first listening location or the mean value of the sound
pressure levels calculated for at least two listening location.
8. The method of claim 7, where the cost function is calculated as
the sum of the absolute differences of each calculated sound
pressure level and the reference sound pressure level for each
phase value and each frequency.
9. The method of claim 2, where the cost function is weighted with
a frequency dependent factor that is inversely proportional to the
mean sound pressure level.
10. The method of claim 1, comprising a third loudspeaker having a
third supply channel arranged upstream thereto which comprises a
phase shifter that modifies the phase of the audio signal
transmitted therethrough according to a third phase function, the
method further comprising: calculating a further optimal offset
phase function based on a mathematical model using the estimated
transfer characteristics; updating the further phase function by
superposing the further optimal offset phase function thereto.
11. The method of claim 10, where the phase shifter comprises a
phase filter having filter coefficients defining a phase
response.
12. The method of claim 11 where the phase filter is a finite
impulse response filter, the step of updating the phase function
further comprises: calculating updated filter coefficient values
such that the resulting phase response at least approximately
matches the optimal phase function; and setting the filter
coefficients to the updated filter coefficient values.
13. A system for adapting sound pressure levels in at least one
listening location, comprising: a first loudspeaker and a second
loudspeaker each for generating an acoustic sound signal from an
audio signal; a supply channel arranged upstream to each
loudspeaker receiving the audio signal, the supply channel linked
to the second loudspeaker comprising means for modifying the phase
of the audio signal transmitted therethrough according to a phase
function; means for measuring the acoustic sound signal at each
listening location and providing corresponding electrical signals
representing the measured acoustic sound signal; processing means
for estimating updated transfer characteristics for each pair of
loudspeaker and listening location, for means for calculating based
on a mathematical model using the estimated transfer
characteristics, and for updating the phase function by superposing
the optimal offset phase function thereto.
14. The system of claim 13, where the means for calculating an
optimum offset phase function comprises: means for simulating sound
pressure levels at each listening location for different
frequencies and phase shifts in the supply channel of the second
loudspeaker, where the phase shifts of the audio signals supplied
to the other loudspeakers are initially zero or constant; means for
evaluating a cost function dependent on the sound pressure level
for the different frequencies and phase shifts; and means for
searching a frequency dependent optimal phase shift that yields an
extremum of the cost function, thus obtaining a phase function
representing the optimal phase shift as a function of frequency.
Description
1. CLAIM OF PRIORITY
[0001] This patent application claims priority to European Patent
Application serial number 08 003 731.0 filed on Feb. 28, 2008.
2. FIELD OF THE INVENTION
[0002] The present invention relates to equalizing the sound
pressure level in the low frequency (bass) range generated by a
sound system.
3. RELATED ART
[0003] It is usual practice to manually acoustically optimize
dedicated audio systems, for example in motor vehicles. Although
there have been major efforts to automate this manual process,
these methods and systems, have shown weaknesses in practice or are
extremely complex and costly. In small, highly reflective areas,
such as the interior of a vehicle, poor improvements in the
acoustics are achieved. In some cases, the results are even
worse.
[0004] Especially in the frequency range below approximately 100
Hertz standing waves in the interior of small highly reflective
rooms can cause strongly different sound pressure levels (SPL) in
different listening locations that are, for example, the two front
passenger's seats and the two rear passenger's seats in a motor
vehicle. These different sound pressure levels entail the audio
perception of a person being dependent on his/her listening
location. However, it has been proven by the work of professional
acousticians that it is possible to achieve a good acoustic results
even with relatively simple audio systems.
[0005] A method is known which allows acoustics to be modeled in
virtually any area. However, this so-called wave-field synthesis
requires extensive resources such as computational power, memories,
loudspeakers, amplifier channels, etc. This technique is thus not
suitable for many applications for cost and feasibility reasons,
especially in the automotive industry.
[0006] There is a need for an automatic bass management that is
adequate to replace the previously used, complex process of manual
equalizing by experienced acousticians and that reliably provides
frequency responses in the bass frequency range at predetermined
listening locations which match the profile of predetermined target
functions. Furthermore, it is desirable that a bass management
system be capable to successively adapt the frequency responses in
response to variations of the acoustic properties of the listening
room during operation.
SUMMARY OF THE INVENTION
[0007] A method for adapting sound pressure levels in at least one
listening location, includes generating sound pressure by a first
and a second loudspeaker, each loudspeaker having a supply channel
arranged upstream thereto, where at least the supply channel of the
second loudspeaker modifies the phase of an audio signal
transmitted therethrough according to a phase function. The method
also includes supplying an audio signal to the supply channels and
thus generating an acoustic sound signal; measuring the acoustic
sound signal at each listening location and providing corresponding
electrical signals representing the measured acoustic sound signal;
estimating updated transfer characteristics for each pair of
loudspeaker and listening location; calculating a phase offset
phase function based on a mathematical model using the estimated
transfer characteristics; and updating the phase function by
superposing the optimal offset phase function thereto.
DESCRIPTION OF THE DRAWINGS
[0008] The invention can be better understood with reference to the
following drawings and description. The components in the figures
are not necessarily to scale, instead emphasis being placed upon
illustrating the principles of the invention. Moreover, in the
figures, like reference numerals designate corresponding parts. In
the drawings:
[0009] FIG. 1 illustrates the sound pressure level in decibel over
frequency measured on four different listening locations within a
passenger compartment of a car with an unmodified audio signal
being supplied to the loudspeakers;
[0010] FIG. 2 illustrates standing acoustic waves within the
passenger compartment of a car which are responsible for large
differences in sound pressure level (SPL) between the listening
locations;
[0011] FIG. 3 illustrates an adaptive bass management system;
[0012] FIG. 4 illustrates the sound pressure level in decibel over
phase shift which the audio signal supplied to one of the
loudspeakers is subjected to; a minimum distance between the sound
pressure levels at the listening locations and a reference sound
pressure level is found at the minimum of a cost function
representing the distance;
[0013] FIG. 5 is a plot of the cost function over phase at
different frequencies;
[0014] FIG. 6 illustrates a phase function of optimum phase shifts
over frequency that minimizes the cost function at each frequency
value;
[0015] FIG. 7 illustrates the approximation of the phase function
by the phase response of a 4096 tap FIR all-pass filter; and
[0016] FIG. 8 illustrates the performance of the FIR all-pass
filter of FIG. 7 and the effect on the sound pressure levels at the
different listening locations.
DETAILED DESCRIPTION
[0017] While reproducing an audio signal using a loudspeakers or a
set of loudspeakers in a car, measurements in the passenger
compartment of the car yield considerably different results for the
sound pressure level (SPL) observed at different listening
locations even if the loudspeakers are symmetrically arranged
within the car. The diagram of FIG. 1 illustrates this effect. In
the diagram four curves are depicted, each illustrating the sound
pressure level in decibel (dB) over frequency which have been
measured at four different listening locations in the passenger
compartment, namely near the head restraints of the two front and
the two rear passenger seats, while supplying an audio signal to
the loudspeakers. One can see that the sound pressure level
measured at listening locations in the front of the room and the
sound pressure level measured at listening locations in the rear
differ by up to 15 dB dependent on the considered frequency.
However, the biggest gap between the SPL curves can be typically
observed within a frequency range from approximately 40 to 90 Hertz
which is part of the bass frequency range.
[0018] "Bass frequency range" is not a well-defined term but widely
used in acoustics for low frequencies in the range from, for
example, 0 to 80 Hertz, 0 to 120 Hertz or even 0 to 150 Hertz.
Especially when using car sound systems with a subwoofer placed in
the rear window shelf or in the rear trunk, an unfavorable
distribution of sound pressure level within the listening room can
be observed. The SPL maximum between 60 and 70 Hertz (cf. FIG. 1)
may likely be regarded as booming and unpleasant by rear
passengers.
[0019] The frequency range where a big discrepancy between the
sound pressure levels in different listening locations, especially
between locations in the front and in the rear of the car, can be
observed depends on the dimensions of the listening room. The
reason for this will be explained with reference to FIG. 2 which is
a schematic side-view of a car. A half wavelength (denoted as
.lamda./2) fits lengthwise in the passenger compartment. A typical
length of .lamda./2=2.5 m yields a frequency of f=c/.lamda.=68 Hz
when assuming a speed of sound of c is equal to 340 m/s. It can be
seen from FIG. 1, that approximately at this frequency a maximum
SPL can be observed at the rear listening locations. Therefore it
can be concluded that superpositions of several standing waves in
longitudinal and in lateral direction in the interior of the car
(the listening room) are responsible for the inhomogeneous SPL
distribution in the listening room.
[0020] In order to achieve more similar--in the best case
equal--SPL curves (magnitude over frequency) at a given set of
listening locations within the listening room an automatic
equalization of the sound pressure levels is suggested and
explained below by way of examples. For the following discussion it
is assumed that only two loudspeakers are arranged in a listening
room (e.g., a passenger compartment of a car) where four different
listening locations are of interest, namely a front left (FL), a
front right (FR) a rear left (RL) and a rear right (RR) position.
Of course the number of loudspeakers and listening locations is not
limited. The method may be generalized to an arbitrary number of
loudspeakers and listening locations. FIG. 3 illustrates such an
audio system comprising two loudspeakers 20a, 20b and four
listening positions (FL, FR, RL, RR) where a microphone 10a, 10b,
10c, 10d is provided at each listening location.
[0021] Both loudspeakers 20a, 20b are supplied with the same audio
signal via supply channels (i.e., output channels of the signal
source) comprising amplifiers 30a, 30b. Consequently both
loudspeakers 20a, 20b contribute to the generation of the
respective sound pressure level in each listening location. The
audio signal is provided by a signal source 50 having an output
channel for each loudspeaker to be connected. At least the output
channel supplying the second one of the loudspeakers 20a, 20b is
configured to apply a programmable phase shift .phi.(f) to the
audio signal supplied to the second loudspeaker. The phase shift
.phi.(f) is provided by a phase filter 40, for example, an FIR
all-pass. A processing unit 60 calculates filter coefficients for
the phase filter 40 from measured sound pressure levels SPL.sub.FL,
SPL.sub.FR, SPL.sub.RL, SPL.sub.RR received from the microphones
10a, 10b, 10c, and 10d respectively. For calculating the filter
coefficients of the phase filter 40 a predefined target function
may be considered, that is, the filter coefficients are adapted
such that the frequency responses of the sound pressure levels
SPL.sub.FL(f), SPL.sub.FR(f), SPL.sub.RL(f), SPL.sub.RR(f) at the
listening locations approximate the predefined target function
SPL.sub.REF(f). The functionality provided by the processing unit
60 is explained in the further discussion, that is, the processing
unit is configured to perform at least one of the methods explained
below.
[0022] The sound pressure level observed at a listening locations
of interest will change dependent on the phase shift applied to the
audio signal that is fed to the second loudspeaker 20b, while the
first loudspeaker 20a receives the same audio signal with no phase
shift applied to it. Of course the audio signal supplied to the
first loudspeaker 20a may also be phase shifted, but only the
relative phase shifts between the considered audio signals is
relevant. Consequently, the phase shift of the audio signal
supplied to the first loudspeaker 20a may be arbitrarily set to
zero for the following discussion. The dependency of sound pressure
level SPL in decibel (dB) on phase shift .phi. in degree (.degree.)
at a given frequency f (in this example 70 Hz) is illustrated in
FIG. 4 as well as the mean level of the four sound pressure levels
measured at the four different listening locations.
[0023] A cost function CF(.phi.) is provided which represents the
"distance" between the four sound pressure levels
SPL.sub.FL(.phi.), SPL.sub.FR(.phi.), SPL.sub.RL(.phi.),
SPL.sub.RR(.phi.) and a reference sound pressure level
SPL.sub.REF(.phi.) at a given frequency f. Such a cost function may
be defined as:
CF(.phi.)=|SPL.sub.FL(.phi.)-SPL.sub.REF(.phi.)|+|SPL.sub.FR(.phi.)-SPL.-
sub.REF(.phi.)|+|SPL.sub.RL(.phi.)-SPL.sub.REF(.phi.)|+|SPL.sub.RR(.phi.)--
SPL.sub.REF(.phi.)|, (1)
where the symbols SPL.sub.FL, SPL.sub.FR, SPL.sub.RL, SPL.sub.RR
denote the sound pressure levels at the front left, the front
right, the rear left and the rear right positions respectively. The
symbol .phi. in parentheses indicate that each sound pressure level
is a function of the phase shift .phi.. The distance between the
actually measured sound pressure level and the reference sound
pressure level SPL.sub.REF is a measure of quality of equalization,
that is, the lower the distance, the better the actual sound
pressure level approximates the reference sound pressure level. In
the case that only one listening location is considered, the
distance may be calculated as the absolute difference between
measured sound pressure level and reference sound pressure level
SPL.sub.REF, which may theoretically become zero.
[0024] Equation 1 is an example for a cost function whose function
value becomes smaller as the sound pressure levels SPL.sub.FL,
SPL.sub.FR, SPL.sub.RL, SPL.sub.RR approach the reference sound
pressure level SPL.sub.REF. At a given frequency, the phase shift
.phi. that minimizes the cost function yields an "optimum"
distribution of sound pressure level, that is, the sound pressure
level measured at the four listening locations have approached the
reference sound pressure level SPL.sub.REF as good as possible and
thus the sound pressure levels at the four different listening
locations are equalized resulting in an improved room acoustics. In
the example of FIG. 4, the mean sound pressure level is used as
reference SPL.sub.REF and the optimum phase shift that minimizes
the cost function CF(.phi.) has been determined to be approximately
180.degree. (indicated by the vertical line).
[0025] The cost function may be weighted with a frequency dependent
factor that is inversely proportional to the mean sound pressure
level. Accordingly, the value of the cost function is weighted less
at high sound pressure levels. As a result an additional
maximization of the sound pressure level can be achieved. Generally
the cost function may depend on the sound pressure level, and/or
the above-mentioned distance and/or a maximum sound pressure level.
Furthermore, the reference SPL.sub.REF is not necessarily the mean
sound pressure level as in Equation 1. The front left sound
pressure level SPL.sub.FL may also be used as a reference sound
pressure level SPL.sub.REF as well as a predefined target function.
In the latter case the reference sound pressure level SPL.sub.REF
is not dependent on the phase shift .phi., but only a function of
frequency.
[0026] In the above example, the optimal phase shift has been
determined to be approximately 180.degree. at a frequency of the
audio signal of 70 Hz. Of course the optimal phase shift is
different at different frequencies. Defining a reference sound
pressure level SPL.sub.REF(.phi., f) for every frequency of
interest allows for defining cost function CF(.phi., f) being
dependent on phase shift and frequency of the audio signal. An
example of a cost function CF(.phi., f) being a function of phase
shift and frequency is illustrated as a 3D-plot in FIG. 5. The mean
of the sound pressure level measured in the considered listening
locations may be used as reference sound pressure level
SPL.sub.REF(.phi., f). However, the sound pressure level measured
at a certain listening location or any mean value of sound pressure
levels measured in at least two listening locations may be used.
Alternatively, a predefined target function (frequency response) of
desired sound pressure levels may be used as reference sound
pressure level SPL.sub.REF(f). Combinations of the above examples
may also be useful.
[0027] For each frequency f of interest an optimum phase shift can
be determined by searching the minimum of the respective cost
function as explained above, thus obtaining a phase function of
optimal phase shifts .phi..sub.OPT(f) as a function of frequency.
An example of such a phase function .phi..sub.OPT(f) (derived from
the cost function CF(.phi., f) of FIG. 5) is illustrated in FIG.
6.
[0028] A technique for obtaining such a phase function
.phi..sub.OPT(f) of optimal phase shifts for a sound system having
a first and a second loudspeaker (cf. FIG. 3) shall now be
summarized.
[0029] Supply an audio signal of a programmable frequency f to each
loudspeaker. As explained above, the second loudspeaker has a delay
element (e.g., phase filter) connected upstream thereto to apply a
programmable phase-shift .phi. to the respective audio signal.
[0030] Measure the sound pressure level SPL.sub.FL(.phi., f),
SPL.sub.FR(.phi., f), SPL.sub.RL(.phi., f), SPL.sub.RR(.phi., f) at
each listening location for different phase shifts .phi. within a
certain phase range (e.g., 0.degree. to 360.degree.) and for
different frequencies within a certain frequency range (e.g., 0 Hz
to 150 Hz).
[0031] Calculate the value of a cost function CF(.phi., f) for each
pair of phase shift .phi. and frequency f, where the cost function
CF(.phi., f) is dependent on the sound pressure level
SPL.sub.FL(.phi., f), SPL.sub.FR(.phi., f), SPL.sub.RL(.phi., f),
SPL.sub.RR(.phi., f), and optionally on a target function of
desired sound pressure levels.
[0032] Search, for every frequency value f for which the cost
function has been calculated, the optimal phase shift
.phi..sub.OPT(f) which minimizes the cost function CF(.phi., f),
that is:
CF(.phi..sub.OPT,f)=min{CF(.phi.,f)} for
.phi..epsilon.[0.degree.,360.degree.], (2)
thus obtaining a phase function .phi..sub.OPT(f) representing the
optimal phase shift .phi..sub.OPT(f) as a function of
frequency.
[0033] Of course, in practice the cost function is calculated for
discrete frequencies f=f.sub.k.epsilon.{f.sub.0, f.sub.1, . . . ,
f.sub.K-1} and for discrete phase shifts
.phi.=.phi..sub.n.epsilon.{.phi..sub.0, .phi..sub.1, . . . ,
.phi..sub.N-1}, where the frequencies may be a sequence of discrete
frequencies with a fixed step-width .DELTA.f (e.g., .DELTA.f=1 Hz)
as well as the phase shifts may be a sequence of discrete phase
shifts with a fixed step-width .DELTA..phi. (e.g.
.DELTA..phi.=1.degree.). In this case the calculated values of the
cost function CF(.phi., f) may be arranged in a matrix CF[n, k]
with lines and columns, where a line index k represents the
frequency f.sub.k and the column index n the phase shift
.phi..sub.n. The phase function .phi..sub.OPT(f.sub.k) can then be
found by searching the minimum value for each line of the matrix.
In mathematical terms:
.PHI. OPT ( f k ) = .PHI. i for CF [ i , k ] = min { CF [ n , k ] }
, n .di-elect cons. { 0 , N - 1 } , k .di-elect cons. { 0 , K - 1 }
. ( 3 ) ##EQU00001##
[0034] For an optimum performance of the bass reproduction of the
sound system the optimal phase shift .phi..sub.OPT(f), which is to
be applied to the audio signal supplied to the second loudspeaker,
is different for every frequency value f. A frequency dependent
phase shift may be implemented by an all-pass filter (cf. phase
filter 40 of FIG. 3) whose phase response has to be designed to
match the phase function .phi..sub.OPT(f) of optimal phase shifts
as good as possible. An all-pass with an phase response equal to
the phase function .phi..sub.OPT(f) that is obtained as explained
above would equalize the bass reproduction in an optimum manner. A
FIR all-pass filter may be appropriate for this purpose although
some trade-offs have to be accepted. In the following examples a
4096 tap FIR-filter is used for implementing the phase function
.phi..sub.OPT(f). However, Infinite Impulse Response (IIR)
filters--or so-called all-pass filter chains--may also be used
instead, as well as analog filters, which may be implemented as
operational amplifier circuits.
[0035] Referring to FIG. 6, one can see that the phase function
.phi..sub.OPT(f) comprises many discontinuities resulting in very
steep slopes d.phi..sub.OPT/df. Such steep slopes d.phi..sub.OPT/df
may be implemented by FIR filters with a sufficient precision when
using extremely high filter orders, which is problematic in
practice. Therefore, the slope of the phase function
.phi..sub.OPT(f) is limited, for example, to .+-.10.degree.. This
means that the minimum search (cf. Equation 3) is performed with
the constraint (side condition) that the phase must not differ by
more than 10.degree. per Hz from the optimum phase determined for
the previous frequency value. In mathematical terms, the minimum
search is performed according equation 3 with the constraint:
|.phi..sub.OPT(f.sub.k)-.phi..sub.OPT(f.sub.k-1)|/|f.sub.k-f.sub.k-1|<-
;10.degree.. (4)
[0036] In other words, in the present example the function "min"
(cf. equation 3) does not just mean "find the minimum" but "find
the minimum for which equation 4 is valid". In practice the search
interval where the minimum search is performed is restricted.
[0037] FIG. 7 is a diagram illustrating a phase function
.phi..sub.OPT(f) obtained according to Equations 3 and 4 where the
slope of the phase has been limited to 10.degree./Hz. The phase
response of a 4096 tap FIR filter that approximates the phase
function .phi..sub.OPT(f) is also depicted in FIG. 7. The
approximation of the phase is regarded as sufficient in practice.
The performance of the FIR all-pass filter compared to the "ideal"
phase shift .phi..sub.OPT(f) is illustrated in FIGS. 8a and 8d.
[0038] The examples described above comprise SPL measurements in at
least two listening locations. However, for some applications it
might be sufficient to determine the SPL curves only for one
listening location. In this case a homogenous SPL distribution
cannot be achieved, but with an appropriate cost function an
optimization in view of another criterion may be achieved. For
example, the achievable SPL output may be maximized and/or the
frequency response, that is, the SPL curve over frequency, may be
"designed" to approximately fit a given desired frequency response.
Thereby the tonality of the listening room can be adjusted or
"equalized", which is a common term used therefore in
acoustics.
[0039] As described above, the sound pressure levels at each
listening location may be actually measured at different
frequencies and for various phase shifts. However, this
measurements alternatively may be fully or partially replaced by a
model calculation to determine the sought SPL curves by simulation.
For calculating sound pressure level at a defined listening
location knowledge about the transfer characteristic from each
loudspeaker (cf. loudspeakers 20a, 20b in FIG. 3) to each listening
location (cf. locations FL, FR, RL, RR in FIG. 3) is required. In
the case of the system of FIG. 3 (four listening locations and two
loudspeakers) eight transfer characteristics, for example,
frequency or impulse responses, have to be determined.
[0040] Consequently, before starting calculations the overall
transfer characteristic from the loudspeakers to the listening
locations have to be identified, for example, estimated from
measurements. For example, the impulse responses may be estimated
from sound pressure level measurements when supplying a broad band
signal consecutively to each loudspeaker. In addition, adaptive
filters may be used for estimation. Other known methods for
parametric and nonparametric model estimation may also be
employed.
[0041] After the necessary transfer characteristics have been
determined, the desired SPL curves, for example the matrix
visualized in FIG. 4, may be calculated based on a model, that is,
based on the previously determined transfer characteristics.
Thereby one transfer characteristic, for example an impulse
response, is associated with a certain pair of loudspeaker and
listening location. The sound pressure level is calculated by
simulation at each listening location assuming, for the
calculation, that a simulated audio signal of a programmable
frequency is supplied to each loudspeaker, where the audio signal
supplied to the second loudspeaker is phase-shifted by a
programmable phase shift relatively to the simulated audio signal
supplied to the first loudspeaker. Thereby, the phase shifts of the
audio signals supplied to the other loudspeakers are initially zero
or constant. In this context the term "assuming" has to be
understood considering the mathematical context, that is, the
frequency, amplitude and phase of the audio signal are used as
input parameters in the model calculation. In other words, the
above described measurements of sound pressure levels at different
frequencies and phase shifts may be simulated.
[0042] For each listening location this model based calculation may
be split up in the following steps where the second loudspeaker has
a phase-shifting element with the programmable phase shift
connected upstream thereto:
[0043] Calculate amplitude and phase of the sound pressure level
generated by the first and the second loudspeaker, alternatively by
all loudspeakers, at the considered listening location when
supplied with an audio signal of a frequency f using the
corresponding transfer characteristics (e.g., impulse responses)
for the calculation, whereby the second loudspeaker is assumed to
be supplied with an audio signal phase shifted by an amount .phi.
with respect to the audio signal supplied to the first
loudspeaker;
[0044] Superpose with proper phase relation the above calculated
sound pressure levels thus obtaining a total sound pressure level
at the considered listening location as a function of frequency f
and phase shift .phi..
[0045] Once the SPL curves for the relevant phase and frequency
values have been calculated, the optimal phase shift for each
considered loudspeaker may be determined as described above. The
effect of the phase shift may be subsequently determined for each
further loudspeaker.
[0046] In the examples presented above, a system comprising only
two loudspeakers and four listening locations of interest has been
assumed. In such a system only one optimal phase function has to be
determined and the corresponding FIR filter implemented in the
channel supplying one of the loudspeakers (referred to as second
loudspeaker in the above examples). In a system with more than two
loudspeakers, an additional phase function of optimal phase shifts
.phi..sub.OPTi (index i denotes the respective loudspeaker) has to
be determined and a corresponding FIR all-pass filter has to be
implemented in the channel supplying each additional loudspeaker.
If more than four listening locations are of interest all of them
have to be considered in the respective cost function. A more
general approach may be summarized as follows:
[0047] (a) perform the following steps for each of the L
loudspeakers i=2, 3, . . . , L;
[0048] (b) determine the transfer characteristic of each
combination of the loudspeaker and listening locations;
[0049] (c) simulate, using the transfer characteristics, for
different frequencies and different phase shifts of the audio
signal related to the considered loudspeaker, the sound pressure
level at each listening location, where the phase shifts of the
audio signals supplied to the other loudspeakers are initially zero
or constant;
[0050] (d) calculate, for pairs of phase shifts and frequencies, a
cost function dependent on the calculated sound pressure levels;
and
[0051] (e) search a frequency dependent optimal phase shift that
yields an extremum (e.g., optimum) of the cost function, thus
obtaining a phase function representing the optimal phase shift as
a function of frequency; and
[0052] (f) set coefficients of a phase filter upstream to the
considered loudspeaker to provide a phase response that at least
approximately matches the phase function of optimal phase
shifts.
[0053] As explained later in more detail, the above-described
method can also be employed to determine an optimal offset phase
function .DELTA..phi..sub.OPT(f) for correcting an initial phase
function .DELTA..phi..sub.OPT(f) previously imposed to the signal
path of a loudspeaker.
[0054] For an adaptive bass management the estimated transfer
characteristics have to be updated in order to allow for
accommodating to slowly varying transfer characteristics during
operation of the audio system. At the end of the production
process, the listening room (e.g., the interior of a car) may be
equipped with an audio system comprising a bass management system
and the above-mentioned transfer characteristics may then be
identified using one of the methods discussed above. These transfer
characteristics are stored in a memory of the audio system and used
as initial transfer characteristics for the subsequent adaptation
process during normal operation of the audio system.
[0055] In adaptive bass management variations of the transfer
characteristics from the loudspeakers 20a, 20b to the listening
locations FL, FR, RL, RR are considered (cf. FIG. 3). This is done
by regularly updating the estimated impulse responses (respectively
transfer functions) during operation starting from a-priori known
initial transfer characteristics that may be determined after the
installation of the audio system.
[0056] In each adaptation step updated transfer characteristics
from the loudspeakers 20a, 20b to each microphone 10a, 10b, 10c,
10d are calculated considering the filter 40 (cf. FIG. 3) providing
a certain phase response .phi..sub.k(f). The filter is arranged in
a signal path (output channel) upstream to a given loudspeaker
(e.g., loudspeaker 20b). The index k represents the number of the
adaptation step. The changes of the room transfer functions between
the loudspeakers and the microphones happen slowly, hence we can
assume the impulse responses as constant, for a certain time
interval. Within this time interval, an optimal offset phase
function .DELTA..phi..sub.OPT(f) may be calculated for each
considered frequency employing the purely model based method, as
described above. After the calculation of the optimal offset phase
function .DELTA..phi..sub.OPT(f) an updated phase function
.phi..sub.k+1(f) (ideal phase response of the phase filter 40) may
be calculated:
.phi..sub.k+1(f)=.phi..sub.k(f)+.DELTA..phi..sub.OPT(f).
[0057] A new set of (approximated) filter coefficients may then be
calculated from the phase function as already described with
reference to the methods discussed before. The adaptive bass
management system works properly if the bandwidth of the reproduced
audio signal during operation has enough signal power in the
considered bass frequency range (e.g., 20 Hz to 150 Hz) to allow
for a proper estimation of the required updated transfer
characteristics.
[0058] The procedure may be repeated permanently during operation
of the audio system. The bass management system is then capable to
adapt to varying environmental conditions that lead to changes in
the transfer characteristics from the loudspeakers to the listening
locations.
[0059] As explained above, transfer characteristics from each
single loudspeaker to each listening location are required for a
proper model based calculation of the optimal phase function
.phi..sub.OPT(f) or the optimal offset phase function
.DELTA..phi..sub.OPT(f), respectively. During normal operation of
the audio system, an acoustic sound signal (e.g., music signal) is
simultaneously radiated from all loudspeakers which makes it
difficult to find an updated transfer characteristics for each
single pair of loudspeaker and listening location. However,
starting from an a-priori known transfer characteristic (which once
has been previously determined) certain mathematical algorithms may
be used for calculating the desired updated transfer
characteristics from measurements of overall transfer functions
describing the transfer characteristics from all loudspeakers to
each considered listening location. Such algorithms may, for
example, be multiple-error least-mean-square (MELMS)
algorithms.
[0060] When reproducing stereo sound, or surround sound
(multi-channel audio) like DTS 5.1 discrete, Dolby digital 5.1,
etc., the audio channels may be monitored, and, if a time interval
is detected where only one loudspeaker is active, the corresponding
transfer characteristics for this single loudspeaker are
determined. The occurrence of such time intervals depends on the
sound (music) signal actually reproduced. In this way the transfer
characteristics may be estimated separately for each loudspeaker
instead of overall transfer characteristics. When estimating a
transfer characteristic from one single loudspeaker to one certain
listening location the other loudspeakers do not necessarily have
to be silent, but the signal levels (volume) of the other
loudspeakers have to be sufficiently silent or the signals radiated
from the other loudspeakers have to be uncorrelated to the signal
radiated from the considered loudspeaker. In the latter case the
signals of the other loudspeakers may be treated as noise. However,
an increased noise level due to the other loudspeaker signals
(being uncorrelated with the considered loudspeaker signal) has a
negative impact on the quality of estimation of the sought transfer
characteristics. The best performance of the estimation is achieved
if only the considered loudspeaker is active during measurements
used for estimation of the sought transfer characteristics.
[0061] Once having estimated updated transfer characteristics for
each pair of loudspeaker and listening location, the adaptation
method may continue as described above and discussed below in more
detail.
[0062] One example of the adaptive technique for setting optimal
phase shift values .phi..sub.k+1(f) by adding optimal phase shift
offset .DELTA..phi..sub.OPT(f) to the actual phase shift values
.phi..sub.k(f) in the signal path of a loudspeaker during operation
of the audio system is now summarized on the basis of the exemplary
audio system of FIG. 3 having four listening locations FL, FR, RL,
RR and two loudspeakers 20a, 20b:
[0063] (a) reproduce an audio signal via at least two signal paths
each supplying a loudspeaker 20a, 20b generating an acoustic sound
signal; the audio signal comprises signal components that cover at
least the bass range, for example the frequency range from 20 Hz to
150 Hz; one signal path (e.g., the one supplying loudspeaker 20b)
comprises a phase shifter 40 that provides a phase shift
.phi..sub.k(f) to the signal being supplied to the respective
loudspeaker 20b, whereas the phase shift imposed to the other
signal path is zero or constant; initial transfer characteristics
of each pair of loudspeaker and listening location being a-priori
known from separate measurements;
[0064] (b) receive the resulting sound signal, at each listening
location FL, FR, RL, RR, and provide electrical signals
representing the sound signal at the respective listening
location;
[0065] (c) estimate updated transfer characteristics (e.g., impulse
response or frequency response) for each pair of loudspeaker (20a,
20b) and listening location (FL, FR, RL, RR) from the electrical
signals and the audio signal;
[0066] (d) calculate the frequency dependent phase shift offset
.DELTA..phi..sub.OPT(f) based on a model;
[0067] (e) update the phase shift .phi..sub.k(f) to the audio
signal supplying the second loudspeaker 20b according to the
equation
.phi..sub.k+1(f)=.phi..sub.k(f)+.DELTA..phi..sub.OPT(f).
[0068] (f) perform the subsequent adaptation step by repeating the
above steps with an updated phase shift .phi..sub.k+1(f).
[0069] If more than two loudspeakers are used the steps (a) to (f)
of the above method may be repeated for all loudspeakers except the
first one.
[0070] The SPL curves depicted in the diagrams of FIG. 8 have been
obtained by simulation to demonstrate the effectiveness of the
method described above. FIG. 8a illustrates the sound pressure
levels SPL.sub.FL, SPL.sub.FR, SPL.sub.RL, SPL.sub.RR measured at
the four listening locations before equalization, that is, without
phase modifications applied to the audio signal. The thick black
solid line represents the mean of the four SPL curves. The mean SPL
has also been used as reference sound pressure level SPL.sub.REF
for equalization. As in FIG. 1 a large discrepancy between the SPL
curves is observable, especially in the frequency range from 40 to
90 Hz.
[0071] FIG. 8b illustrates the sound pressure levels SPL.sub.FL,
SPL.sub.FR, SPL.sub.RL, SPL.sub.RR measured at the four listening
locations after equalization using the optimal phase function
.phi..sub.OPT(f) of FIG. 6 (without limiting the slope
.phi..sub.OPT/df). One can see that the SPL curves are much more
alike (i.e., equalized) and deviate by small amounts from the mean
sound pressure level (thick black solid line).
[0072] FIG. 8c illustrates the sound pressure levels SPL.sub.FL,
SPL.sub.FR, SPL.sub.RL, SPL.sub.RR measured at the four listening
locations after equalization using the slope-limited phase function
of FIG. 7. It is noteworthy that the equalization performs almost
as good as the equalization using the phase function of FIG. 6. As
a result the limitation of the phase change to approximately
10.degree./Hz is regarded as a useful measure that facilitates the
design of a FIR filter for approximating the phase function
.phi..sub.OPT(f).
[0073] FIG. 8d illustrates the sound pressure levels SPL.sub.FL,
SPL.sub.FR, SPL.sub.RL, SPL.sub.RR measured at the four listening
locations after equalization using a 4096-tap FIR all-pass filter
for providing the necessary phase shift to the audio signal
supplied to the second loudspeaker. The phase response of the FIR
filter is depicted in the diagram of FIG. 7. The result is also
satisfactory. The large discrepancies occurring in the unequalized
system are avoided and acoustics of the room is substantially
improved.
[0074] In the examples presented above, a system comprising only
two loudspeakers and four listening locations of interest has been
assumed. In such a system only one optimal phase function has to be
determined and the corresponding FIR filter implemented in the
output channel (i.e., signal path) supplying one of the
loudspeakers (referred to as second loudspeaker in the above
examples). In a system with more than two loudspeakers an
additional phase function has to be determined and a corresponding
FIR all-pass filter has to be implemented in the output channel
supplying each additional loudspeaker. If more than four listening
locations are of interest all of them have to be considered in the
respective cost function. The general procedure of adaptive bass
management may be summarized as follows:
[0075] (a) assign a number i=1, 2, . . . , L to each one of L
loudspeakers and the corresponding output channels.
[0076] (b) Supply a broad band audio signal (e.g., a music signal)
via L signal paths (output channels) to each loudspeaker 1, 2, . .
. , L. Loudspeakers 1 to L receive the respective audio signal from
a signal source which has one output channel per loudspeaker
connected thereto. At least the channels supplying loudspeakers 2
to L modify the phase .phi..sub.2,k(f), .phi..sub.3,k(f), . . . ,
.phi..sub.L,k(f) of the respective audio signal according to
predetermined phase functions (phase .phi..sub.1(f) may be zero or
constant); an acoustic sound signal is thus radiated by the
loudspeakers 1 to L during the adaptation method; initial transfer
characteristics of each pair of loudspeaker and listening location
being a-priori known from separate measurements.
[0077] (c) Receive the resulting sound signal, at each listening
location FL, FR, RL, RR, and provide electrical signals
representing the sound signal at the respective listening
location.
[0078] (d) Estimate updated transfer characteristics for each pair
of the loudspeaker (1, 2, . . . , L) and listening location (FL,
FR, RL, RR) from the respective electrical signals, the audio
signal and the initial transfer characteristics.
[0079] (e) Calculate, for loudspeaker number i=2, the frequency
dependent optimal phase shift offset .DELTA..phi..sub.OPT2(f) based
on a model using the updated transfer characteristics as explained
above.
[0080] (f) Update the phase shifter that modifies the phase
upstream of loudspeaker number i=2, in order to (at least
approximately) provide an updated phase shift
.phi..sub.2,k+1(f)=.phi..sub.2,k(f)+.DELTA..phi..sub.OPT2(f).
[0081] (g) Repeat steps (a) and (f) for loudspeakers i=3, . . . ,
L, thus obtaining updated phase shifts .phi..sub.3,k+1(f), . . . ,
.phi..sub.L,k+1(f).
[0082] (h) Continue the adaptation process by repeating the above
steps (c) to (g), thus subsequently obtaining updated phase shifts
.phi..sub.i,k+2(f), .phi..sub.i,k+3(f), . . . for all loudspeakers
i=2 to L.
[0083] From FIGS. 8b-d it can be seen that a substantial difference
in sound pressure levels was not be equalized in a frequency range
from about 20 to 30 Hz. This is due to the fact that only one
loudspeaker (e.g., the subwoofer) of the sound system under test is
able to reproduce sound with frequencies below 30 Hz. Consequently,
in this frequency range the other loudspeakers were unable to
radiate sound and therefore can not be used for equalizing. If a
second subwoofer would be employed then this gap in the SPL curves
may be "closed", too.
[0084] After equalizing all the loudspeakers as explained above an
additional frequency-dependent gain may be applied to all channels
in order to achieve a desired magnitude response of the sound
pressure levels at the listening locations of interest. This
frequency-dependent gain is the same for all channels.
[0085] The above-described examples relate to techniques for
equalizing sound pressure levels in at least two listening
locations. Thereby a "balancing" of sound pressure is achieved.
However, the method can be also usefully employed when the
"balancing" is the not goal of optimization, but rather a
maximization of the sound pressure at the listening locations
and/or the adjusting of actual sound pressure curves (SPL over
frequency) to match a "target function". In this case the cost
function has to be chosen accordingly. If only the maximization of
sound pressure or the adjusting of the SPL curve(s) in order to
match a target function is to be achieved, this can also be done
for only one listening location. In contrast, at least two
listening locations have to be considered when a balancing is
desired.
[0086] For a maximization of sound pressure level the cost function
is dependent from the sound pressure level at the considered
listening location. In this case the cost function has to be
maximized in order to maximize the sound pressure level at the
considered listening location(s). Thus the SPL output of an audio
system may be improved in the bass frequency range without
increasing the electrical power output of the respective audio
amplifiers.
[0087] As disclosed above, a first example of a technique for
adapting sound pressure levels in at least one listening location
comprises generating the sound pressure using first and a second
loudspeakers, each loudspeaker having a supply channel arranged
upstream thereto, where at least the supply channel of the second
loudspeaker modifies the phase of an audio signal transmitted
therethrough according to a phase function. The method further
comprises: supplying an audio signal to the supply channels and
thus generating an acoustic sound signal; measuring the acoustic
sound signal at each listening location and providing corresponding
signals (e.g., electrical) representing the measured acoustic sound
signal; estimating updated transfer characteristics for each pair
of loudspeaker and listening location; calculating an optimum
offset phase function based on a mathematical model using the
estimated transfer characteristics; updating the phase function by
superposing the optimal offset phase function thereto.
[0088] According to another example, the calculation of an optimum
offset phase function may comprise: simulating, for different
frequencies and phase shifts in the supply channel of the second
loudspeaker, sound pressure levels at each listening location,
where the phase shifts of the audio signals supplied to the other
loudspeakers are initially zero or constant; evaluating, for the
different frequencies and phase shifts, a cost function dependent
on the sound pressure level; and searching a frequency dependent
optimal phase shift that yields an extremum of the cost function,
thus obtaining a phase function representing the optimal phase
shift as a function of frequency.
[0089] In a further example of the invention in the above methods
sound pressure levels in at least two listening locations are
considered.
[0090] In another example of the invention the cost function is
dependent on the calculated sound pressure levels and a previously
defined target function. In this case the actual sound pressure
levels are equalized to the target function.
[0091] Another example of the invention relates to a system for
adapting sound pressure levels in at least one listening location.
The system comprises: a first and a second loudspeaker for
generating an acoustic sound signal from an audio signal; a supply
channel arranged upstream to each loudspeaker receiving the audio
signal, at least the supply channel linked to the second
loudspeaker comprising means for modifying the phase of the audio
signal transmitted therethrough according to a phase function;
sensors for measuring the acoustic sound signal at each listening
location and providing corresponding electrical signals
representing the measured acoustic sound signal; a processing unit
that estimates updated transfer characteristics for each pair of
loudspeaker and listening location; calculates based on a
mathematical model using the estimated transfer characteristics;
and updates the phase function by superposing the optimal offset
phase function thereto.
[0092] Although various examples to realize the invention have been
disclosed, it will be apparent to those skilled in the art that
various changes and modifications can be made which will achieve
some of the advantages of the invention without departing from the
spirit and scope of the invention. It will be obvious to those
reasonably skilled in the art that other components performing the
same functions may be suitably substituted. Such modifications to
the inventive concept are intended to be covered by the appended
claims. Furthermore the scope of the invention is not limited to
automotive applications but may also be applied in any other
environment, for example in consumer applications like home cinema
or the like and also in cinema and concert halls or the like.
* * * * *